计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (6): 127-129.DOI: 10.3778/j.issn.1002-8331.2010.06.036

• 数据库、信号与信息处理 • 上一篇    下一篇

一种应用复杂网络特征的K-means初始化方法

田生文,王伊蕾,李阿丽   

  1. 鲁东大学 计算机科学与技术学院,山东 烟台 264025
  • 收稿日期:2008-09-01 修回日期:2008-11-13 出版日期:2010-02-21 发布日期:2010-02-21
  • 通讯作者: 田生文

K-means initialization method using properties of complex network

TIAN Sheng-wen,WANG Yi-lei,LI A-li   

  1. College of Computer Science and Technology,Ludong University,Yantai,Shandong 264025,China
  • Received:2008-09-01 Revised:2008-11-13 Online:2010-02-21 Published:2010-02-21
  • Contact: TIAN Sheng-wen

摘要: K-means算法是一种基于划分的聚类算法,具有算法简单且收敛速度快的特点。但该算法的性能依赖于聚类中心的初始位置的选择。拓展了复杂网络的重要特征,针对带有属性的数据对象所构成的数据集,定义了多维属性对象的度、聚集度和聚集系数,选取度和聚集系数高的K个点作为K-means聚类的初始中心点。实验数据表明,改进后的K-means算法较传统的算法具有更高的效率和准确度。

关键词: 聚类, K-means算法, 复杂网络特征, 聚类初始点

Abstract: K-means algorithm is a partition-based clustering algorithm.It is simple and fast to converge,the performance of K-means algorithm depends on that how to choose K samples as the initial cluster centers.This paper develops the properties of complex network,and defines degree,congregated degree and congregated coefficient of objects with feature,and chooses the K nodes whose the degree and congregated coefficient are larger than the others as the initial cluster centers.The experiment shows that the improved K-means clustering algorithm is more efficient than the original K-means clustering algorithm.

Key words: clustering, K-means, complex network characteristics, initial cluster centers

中图分类号: